Monday, March 3, 2003

Spectrum: Commons and Corpses

BoingBoing aka Cory Doctorow has blogged up a storm from the Stanford Law confab on property vs. commons in spectrum, as have others. Still looking for a good summation from a business point of view.

Meanwhile, back at the ranch, things are tough for proprietary wireless vendors. In the last couple of weeks, both Centerpoint and Iospan have taken the big sleep.
3:21:06 PM    


FCC: The worst of both worlds

Previously, I had hoped that the FCC was going to make modification to UNE-P (3d party access to local loop) rules that might not be ideal, but would at least give a predictable scenario for competition. Spoke too soon. A split 3-2 vote has allowed the RBOCs to keep some of their new plant closed, but devolved regulation over unbundling of copper loop to the states, an outcome that appears to be pleasing no one:

"It looks like you made a decision that doesn't please anyone," said Rep. Anna Eshoo, a California Democrat and member of the House Commerce Committee's telecom panel. Committee Chairman Billy Tauzin, R-La., took [FCC Commissioner] Martin to task repeatedly. Tauzin said the rules crafted by Martin would generate more litigation, deter Wall Street from investing in phone companies and extend the telecom industry's slump.
Wow, Eshoo agrees with Tauzin, and I agree with them both. (Pardon me while I step outside to check for flying pigs.) You want to freeze both incumbent telco and innovator investments? You couldn't have picked a better way to fragment the market than devolution.
Already, the Bells have indicated they intend to fight the plan to give states more power on local-phone networks. The Bells are now required to lease access to any portion of their network at deep discounts. The Bells say the rules, known as UNE-P, allow rivals to cherry pick their best customers and give them little incentive to invest in newer network technology. Rivals say the Bells want more leeway to squelch competition.
The pity is, they're both right. Anti-incumbent regulators like those here in California will use the opportunity to bludgeon the Bells and score political points, but the resulting markets will be fragmented and dependent on fiat, not a good recipe for investing in alternatives like wireless technology and equipment. What a maroon.
2:46:20 PM    

Telecoms: An analyst after my own heart

The swan song edition of the Red Herring brings an article on the telecoms debacle by Andrew Odlyzko (linked at his personal site, lest RH's website disappear completely). The money graf, as the journo-bloggers say:

"However, while traffic is growing, there is no sign of willingness to dramatically increase spending. Service providers will have to resign themselves to modest increases in revenues, with growth in data coming at the expense of traditional voice. ....we're now entering a phase in which companies and individuals must learn to use a less expensive and more convenient service in a manner that makes economic sense.
Exactly. While the explosion of backbone demand that touched off the telcom frenzy was real enough - I saw it first hand at CompuServe in 1994-6 - many of the forces that drove it were one-time: increases in modem and other link speeds, shift from text to graphics intensive web, rapidly rising user counts, higher link duty cycles from web surfers. With many of these drivers now played out (at least domestically), the network owners are having to shift from simply filling demand to stimulating it, something difficult to accomplish in their current financial condition.

I've had kind words for Prof. Odylzko before, seeing as he wisely agrees with this writer on the folly of content driven business models. What's not to like about a guy whose vitae include titles like "The 10^22-nd zero of the Riemann zeta function" wedged in between telcoms policy and analysis writings? Get a blog!
1:06:23 PM    


Fingerprint sensing is getting cheaper

Biometrics, in particular fingerprints, got a lot of press after 9/11 and the resulting focus on security. One of the problems up to now has been the expense of the sensors, most of which were built from fingertip sized slices of silicon die, and consequently had to sell for $50 and up. Now there's a new generation of sensors coming to market that will drop the sensor costs dramatically.

The secret to reduced expense for most is reducing the sensor size from a full fingerprint to a few rows over which the finger is 'swiped', and software then reconstructs the image. Those offering this technology include:

  • Fingerprint Cards AB, a small Swedish public company. Offers capacitence based swipe sensors for under $10, down to $6 in quantity 500,000. Lots of details due to Swedish disclosure rules.
  • Atmel's Grenoble unit has a swipe product with thermal sensing.
  • Fujitsu has a subsystem level swipe offering apparently based around capacitative technology licensed from Veridicom.
  • National Semi has sensor created through a licensing deal with private company Validity (apparently no website).
Startup Fidelica takes a very different approach to reducing costs. While keeping with sensor sized to the fingertip, it reduces costs by fabricating on glass substrate, rather than using expensive silicon die.
11:28:16 AM    

Postini / Trend Micro product announcement

Trend Micro today announced the product offering resulting from the Postini relationship cited here in January. Story at ZDNET among other places.
11:14:31 AM